Unintended pregnancy is a serious concern in reproductive health which needs to be addressed. Analyzing 717 pregnant women extracted from the Bangladesh Demographic and Health Survey 1999-2000, this study identified that unwanted pregnancy is significantly associated with higher numbers of living sons (P<0.001), longer marital duration (P<0.05), exceeding desired family size (P<0.001), use of contraception (P<0.01), and breastfeeding practice (P<0.05). For mistimed pregnancy, higher age (P<0.05), breastfeeding practice (P<0.01), exceeding desired family size (P<0.01), pregnancy termination (P<0.05), and having last birth during last three years (P<0.001) were significant. Since unintended pregnancy is strongly associated with exceeding desired family size, a multidimensional approach may be needed through the family planning, health and educational sectors in Bangladesh to maintain desirable family sizes.


Almost all women are at risk for unintended pregnancy throughout the reproductive years (Forrest 1994), which occurs in all socioeconomic strata of the society (Moos 2003; Brown and Eisenberg 1995). About half (48%) of all women aged 15-44 have experienced at least one unintended pregnancy sometime in their reproductive life (Moos 2003; Henshaw 1998). The incidence of unintended pregnancy, which includes both unwanted and mistimed pregnancies (Santelli et al. 2003; Brown and Eisenberg 1995), has long been used as a primary indicator of the state of reproductive health (Trussell et al. 1999). Unwanted pregnancies are those which occurred when the women did not want to have any further pregnancies at all (Brown and Eisenberg 1995). Reasons for unwanted pregnancies are: (1) people's growing desire to have smaller families; (2) the unmet need for family planning; (3) ineffectiveness of contraceptive methods; and (4) unwanted sexual relations (Langer 2002). Women who experience unwanted pregnancy are at a greater risk of complicated pregnancy outcomes, and their children are more likely to experience physical or psychological problems in infancy than those women with wanted pregnancies (Kroelinger and Oths 2000).

In contrast, mistimed pregnancies are those that were wanted by the women at some time, but which occurred sooner than they were wanted. This may happen if (1) women who were hoping to space their births a certain distance apart conceived sooner than was hoped for due to contraceptive failure, and (2) women had intercourse without a contraceptive method despite the fact they did not hope to become pregnant (Brown and Eisenberg 1995).

Each year, globally 40-60 million women seek termination of an unwanted pregnancy under unsafe conditions (Rai and Dali, 2002). The consequences of unintended pregnancy are serious, imposing appreciable burdens on children, women, men, families (Brown and Eisenberg 1995) and their societies (Klima 1998). Both unwanted and mistimed pregnancies are known to be associated with numerous harmful behaviours and outcomes. For example, more than 50% of the unwanted and mistimed pregnancies ended in abortion (Henshaw 1998; Brown and Eisenberg 1995; Forest 1994), which is responsible for at least one in eight maternal deaths worldwide (Rai and Dali 2002). Some of the factors which are known to be associated with unintended pregnancies may include age, being unmarried, low income/economic status/poverty (Henshaw 1998; PRAMS GRAM 1996), maternal education, higher parity, (Gadow et al. 1998), nonuse of contraception at the time becoming pregnant, partner's feeling about the pregnancy, level of dependability and support from partner (Kroelinger and Oths 2000), barriers to accessing contraception, fears about using it, errors in use, lack of backup plans when errors in use occur, elective abortions, late entry to prenatal care, low birth weight, child abuse and neglect, behavioural problems in the offspring (Moos 2003), contraceptive failure, and rape (Klima 1998), among others.

Although unintended pregnancy is a worldwide problem (Klima 1998) and women carrying unintended pregnancies are considered vulnerable to various risk factors leading to poor obstetric and perinatal outcomes (Gadow et al. 1998), the issues of both unwanted and mistimed pregnancies receive little attention in the print and electronic media as compared to teenage pregnancy, nonmarital childbearing and abortion (PRAMS GRAM 1996; Brown and Eisenberg 1995). It is particularly true for Bangladesh, which is one of the most populous countries in the world, and where maternal mortality and morbidity are very high as compared to many other Southeast Asian countries like Bhutan, Indonesia, India, Sri Lanka and Thailand (Sein and Rafei 2002). However, this country has been trying relentlessly to achieve the replacement level of fertility mainly through the family planning programs. As a result the contraceptive prevalence rate has increased significantly, from about 8% in 1975 to about 54% in 1999-2000. The rising trend of contraceptive prevalence rate (CPR) is also associated with the fall in fertility, which declined markedly, from 6.3 births per woman in 1975 to 3.3 births per woman in 1999-2000. Despite such improvements, unplanned pregnancies are still common in Bangladesh. It is reported that if all unwanted births were avoided (which is about 1.1 per woman), the fertility rate in Bangladesh would fall from 3.3 to 2.2 children per woman (Nipport et al. 2001). Therefore, reduction in the rate of unintended pregnancy may be an important strategy to achieve the replacement level (2.1 children per woman) of fertility in Bangladesh. It is also noted that replacement level of fertility can be achieved if the current use of contraception is increased from 54 % to over 70% (Khuda et al. 1999).

Keeping the above-mentioned factors in mind, this article investigated some of the sociodemographic factors that might have some influence over both unwanted and mistimed pregnancies among ever-married women.


A total of 717 pregnant women, extracted from 10,544 currently married women of reproductive age of the Bangladesh Demographic and Health Survey (BDHS) 1999-2000, were identified as eligible for this study. The eligibility criterion was based on the pregnancy status during the survey. The response rate was 97% for women in the survey. The detailed methodology of sampling and data collection is given in the 1999-2000 BDHS report (Nipport et al. 2001).

The pregnant women were extracted using the question: (1) Are you pregnant now? The answer was coded as either yes, no or not sure. If she answered yes, she was considered as pregnant. (2) Further, she was asked to respond the question, "At the time of becoming pregnant, did you want this pregnancy then, did you want to wait until later, or did you not want to have any (more) children at all?" The women who wanted the pregnancy then were considered under the wanted group, who desired pregnancy, but later, were considered under the mistimed group, and who did not want to have any (more) children were considered under the unwanted group.

Since unwanted and mistimed pregnancies should not be grouped into a single unintended category (Pulley et al. 2002), the cases of mistimed (unwanted) pregnancy are excluded to the logistic regression for unwanted (mistimed) pregnancy. The following variables have been used in the logistic regression model.

Dependent variable:
Whether pregnancy wanted or unwanted (if wanted = 0, unwanted = 1)
Whether pregnancy wanted or mistimed (if wanted = 0, mistimed = 1)
Independent variables (categorical):
Age of the women (10-24 years = 0, 25-34 years =1 , 35-49 years = 2)
Level of education (no education = 0, primary education = 1, secondary and above = 1)
Place of residence (urban = 0, rural = 1)
Region of residence (Barisal = 0, Chittagong = 1, Dhaka = 2, Khulna = 3,
Rajshahi = 4, Sylhet = 5)
Religion (non-Muslim = 0, Muslim = 1)
Own television (no = 0, yes = 1)
Marital duration in years (0-9 years = 0, 10-19 years = 1, 20 years and above = 2)
Number of living sons (no son = 0, 1 son = 1, 2 sons or more = 2)
Whether currently breastfeeding (no = 0, yes = 1)
Whether exceeded desire family size with present pregnancy (no = 0, yes = 1)
Whether ever used any contraception (no = 0, yes = 1)
Whether ever had terminated pregnancy (no = 0, yes = 1)
Whether discussed about family planning method with partner in the last year (no = 0, yes = 1)
Whether had given birth in the last 3 years (no = 0, yes = 1)

[Table 1]



Basic Characteristics of Pregnant Women

From Table 1 it is revealed that 62.9% pregnant women belonged to the age category 10-24. Among the respondents 40% had no education, while 31% had primary level of education. Three-fourths of the respondents were from rural areas and 90% were Muslim. Only 19% owned a television (TV). The marital duration was 0-9 years for 69.3% of the women. About one-third of the pregnant women experienced birth during last three years; one-fourth of the respondents exceeded their family size with that pregnancy. Use of contraception was 59.4% among the total women, with 19.3% experiencing terminated pregnancy. About 16% of the women reported breastfeeding.

Differential of Unwanted Pregnancy Rates and χ2 Test

The rate of mistimed and unwanted pregnancy was 27.3% and 14.6% respectively (Table 1). The rate of unwanted pregnancy was highest in the highest age group (55.3% for 35-49), and lowest in the lowest age group (5.1% for 10-24). Education was negatively related with unwanted pregnancy. Marital duration indicated that 62.9% of pregnancy was unwanted when the marital duration was more than 20 years, whereas it was only 4.4% when marital duration was less than or equal to 9 years. Unwanted pregnancy was positively associated with number of living sons. The rate was extraordinarily high (53.0%) when the women had two or more sons as compared to women having one living son (15.9%) only. About 50.0% of the pregnancies were unwanted from those who exceeded the reported number of ideal children. This rate was 20 times higher than those who did not exceed the ideal number of children. About one-fifth of the pregnancies were unwanted for the women who had ever used contraception. Unwanted pregnancy was also positively associated with number of births in the last three years from the date of the survey.

The proportion of unwanted pregnancy was 25.9% for the women who reported breastfeeding at the time of survey, which was double those who were not breastfeeding then. Almost all the variables except place of residence, religion, and husband-wife discussion about family planning matters were significantly associated with pregnancy status by χ2 test.

[Table 2]

The results of the logistic regression analysis are presented in Table 2. Only six variables (education of women, marital duration, number of living sons, whether exceeded the number of ideal children, use of contraception, and currently breastfeeding) were significantly related with unwanted pregnancies. Odds ratios for marital duration indicated that the rate of unwanted pregnancies were 2.7 and 12.0 times higher for the women with marital duration 10-19 years and 20 years and above, as compared to women with marital duration less than or equal to 9 years. The odds ratio was 21 times higher for those who exceeded ideal number of children compared with those who did not exceed the ideal number of children. The ratio was 4.7 times and 2.3 times higher for women with two sons and more and one son respectively than those women without sons. The ratio of unwanted pregnancy was 3.8 times higher for women who had ever used contraception as compared to the reference category. Women who reported breastfeeding had 3.7 times higher odds ratio of unwanted pregnancies than the reference category.

[Table 3]

Table 3 presents the percentage of women who exceeded the reported number of ideal children by some sociodemographic variables. The rate was significantly higher (P<0.001) for women who had no education (35.5%) than the women having education (20.0%). Similarly the rate is significantly higher (P<0.001) for those women who reported current breastfeeding (41.0%), ever used of contraception (32.70%), married for 20 years or more (82.9%), having two sons or more (75.7%) as compared to other categories of each variable.

[Table 4]

Table 4 presents information about the last contraception method used by those who ever used contraception, causes of discontinuation of use in the last five years, and preferred contraceptive method for the future. Though there were 426 (59.4%) "ever" users among the total sample, only 396 responded about last method. The women were classified according to the last contraceptive methods. The Pill was the most common method among the pregnant women for all groups. Condom, injection, periodic abstinence and withdrawal were also reported.

Why did they discontinue contraception use in the last five years? The most frequent answer was that they wanted to become pregnant at that time. Among the unwanted and mistimed groups, important reasons were as follows: became pregnant (which means method failure), side effects, accessibility to methods, and husband disapproval.

Next each pregnant woman was asked to mention the preferable method of contraception which they intended to use in future. In response, 667 women responded; according to their responses, the Pill was the most frequently cited method over others. The answer "don't know" was the second highest, followed by injection and female sterilization respectively.

[Table 5]

Finally an attempt has been made to examine which particular methods were responsible for method failure, side effects and so on (Table 5). The highest rate of method failure was found among periodic abstinence users, followed by withdrawal, condom and pill users respectively. Side effects were found highest among injection users, followed by pill users. All others reasons, such as husband disapproval, health concerns, limited accessibility, and pill users reported so on.


The unwanted pregnancy rate was 15% in Bangladesh among the pregnant women in the survey. Unwanted pregnancy is not unique to Bangladesh, but is common all over the world. For example, unwanted pregnancies were 25% in India (Saha and Chatterjee 1998), 12.5% in Sri Lanka, 14.2% in Thailand, 6.5% in Indonesia (Sein and Rafei 2002), 9% in Zimbabwe (Mbizvo et al. 1997), 20% in Nigeria (Okonofua et al. 1999), 12.5% (PRAMS GRAM 1996) and 4% in USA (Hellerstedt et al. 1998), 37% in South America (Gadow et al. 1998) and 16.5% in Nova Scotia, Canada (Denton and Scot 1994). The substantial variation in unwanted fertility among countries may be caused by the variation in the degree of implementation of preferences, the effectiveness of contraceptive use, the rate of induced abortion, age at marriage, duration of breastfeeding, and frequency of sexual relations (Bongaarts 1997).

Although the χ2 statistic showed many significant associations (Table 1) between pregnancy status and sociodemographic variables, most of them disappeared when logistic regression was applied (Table 2). Only a few of them were significantly associated with unwanted pregnancy. This study confirmed that (i) marital duration, (ii) number of sons, (iii) exceeding desired (ideal) number of children, (iv) ever terminated pregnancy, (v) ever used contraception, and (vi) currently breastfeeding were the significant determinants of unwanted pregnancies in Bangladesh.

The data (not shown) indicated that longer marital duration, higher age and lower educational level of women are represented by higher number of living children or higher number of sons. Since the women with a higher number of living children or number of sons are more likely to exceed the desired number of children, further pregnancy would be more likely to be unintended. Positive association between unintended pregnancy and number of children are also reported by other studies (PRAMS GRAM 1996; Denton and Scott 1994). Higher parity or close pregnancies were positively associated pregnancy termination (Thapa and Padhye 2001) and other abnormal outcome, including maternal death.

Women of low socioeconomic status are at greater risk of contraceptive nonuse and for contraceptive failure; thus they are also at greater risk of unintended conceptions (Forrest 1994). Denton and Scott (1994) reported that contraceptive failure was responsible for 13.5% of unintended pregnancies. However, effective use contraception depends on many factors. Education is one of the most important factors. A woman with higher education is more likely to use contraceptive methods, and is more health-conscious. Education increases the empowerment of women and increases the decision-making power in the family, which is the key to the rest of a woman's development and future (Rao 2001). It also increases the interrelationship between husband and wife, which ultimately helps them to make joint decisions regarding family size, contraceptive use and so on. Educated women have more opportunities (such as access to radio, TV, newspaper) to learn about various contraception, including their advantages and disadvantages. They are also aware of the consequences of large family size. All factors together act to control the family size according to desire level.

Although Taylor and Cabral (2002) and Dye et al. (1997) found an inverse association between unwanted pregnancies and breastfeeding practice, this study failed to support their findings. Surprisingly, the data (Table 1) showed a higher rate of unwanted pregnancy among the women who were breastfeeding their children during the survey. Furthermore, this study showed that women who had ever used contraception were not necessarily less likely to have an unwanted pregnancy. This may be explained by the following reasons: (i) perhaps breastfeeding practice and "ever" use of contraception were not effective in reducing unwanted pregnancy in case of Bangladesh, or (ii) the rate of exceeding the ideal number of children (desire family size) was significantly higher for those women who had ever used contraception and who had been currently breastfeeding (Table 3). Whether women exceeded the ideal number of children may be crucial because it showed the greatest impact on unwanted pregnancy by odds ratio (Table 2). Since a large portion of the women who reported the breastfeeding practice and ever use of contraception had exceeded ideal number of children, perhaps the effectiveness of breastfeeding and ever use of contraception disappeared, and therefore the main factor behind the associations of unwanted pregnancy with breastfeeding and ever use of contraception may be exceeding the ideal number of children, or (iii) these findings could also be the result of confounding factors.

In the case of ever use of contraception, method failure may also contribute in cases of unwanted pregnancy. The failure rate was substantial (18%) among the Pill users, the most common method of contraception in Bangladesh. Many women were practicing periodic abstinence and withdrawal as a contraception method, which showed high rate of method failure (Table 5). It is reported that among the women who became pregnant due to contraceptive failure that 68% defined the pregnancy as unintended (Santelli et al., 2003).

What kind of strategies to reduce the incidence of unwanted pregnancy should be implemented in Bangladesh? First, since strongest association (P<0.001) was found for those women who exceeded desired family size, synergistic efforts of family planning, health and educational programs are an extreme need in Bangladesh to motivate the women for using contraception effectively so that they can limit the number of children according to their desire. As the average number of desired children was around 2.5, which is close to replacement level of fertility, it is not important to counsel all the women to limit their children at 2, but it is very much important to help and motivate women empathetically so that they can not exceed their desired family size.

Second, since shortcomings of contraceptives and family planning delivery systems are major reasons for unwanted pregnancy and unsafe abortion in Third World countries (Kabir 1989), and since family planning services are acknowledged as one of the most cost-effective programs to prevent unwanted pregnancy in all socioeconomic groups (PRAMS GRAM 1996), family planning and health programs should ensure increasing access to the contraception with minimum barriers for all women and should provide all other services to fulfill their multiple reproductive health needs. Higher parity women should be motivated to adopt a permanent method. Special emphasis also should given to the Pill (as a most preferable method) users and others for which failure rate is high.

Third, creating more facilities for women's education may be another contributing factor in reducing unwanted pregnancy. Increasing education of women will help delay their age at first marriage, through which marital duration can be reduced and effective contraceptive use can be increased (method failure can be reduced), and as a result women will have lower parity.

We should remember that in limiting the number of unintended pregnancies, three vital things could be achieved, at least partially: (i) number of children could be limited according to desire, (ii) the gap between actual fertility rate and replacement level of fertility could be minimized, and (iii) the number of hazardous terminated pregnancies or abortions could be reduced. In short, such activities are needed through which women can complete their reproductive age with their desired number of children. Moreover, because of the serious consequences of unwanted pregnancy, every society should adopt a new social norm under which all pregnancies would be intended; i.e., they should be consciously and clearly desired at the time of conception (Brown and Eisenberg 1995).

About the Author

MMH Khan
Department of Public Health, Sapporo Medical University, Japan
Department of Statistics, Jahangirnagar University, Bangladesh

M Kabir
Department of Statistics, Jahangirnagar University, Bangladesh

Mitsuru Mori
Department of Public Health, Sapporo Medical University, Japan


Please send correspondence to MMH Khan, Department of Public Health, Sapporo Medical University School of Medicine, South 1, West 17, Chuo-ku, Sapporo 060-8556, Japan. Email: khan@sapmed.ac.jp, fax: (+81) 11-641-8101.


Bongaarts, J. 1997. "Trends in Unwanted Childbearing in the Developing World." Studies in Family Planning 28: 267-77.

Brown, S.S. and L. Eisenberg. 1995. The Best Intentions: Unintended Pregnancy and the Well-Being of Children and Families. Washington, DC: National Academy Press.

Denton, A.B. and K.E. Scott. 1994. "Unintended and Unwanted Pregnancy in Halifax: The Rate and Associated Factors." Canadian Journal of Public Health 85: 234-38.

Dye, T.D., M. Wojtowycz, R. Aubry , J. Quade and H. Kilburn. 1997. "Unintended Pregnancy and Breast-Feeding Behavior." American Journal of Public Health 87: 1709-11.

Forrest, J.D. 1994. "Epidemiology of Unintended Pregnancy and Contraceptive Use." American Journal of Obstetrics and Gynecology 170: 1485-89.

Gadow, E.C., J.E. Paz, J.S. Lopez-Camelo, M.G. Dutra, J.T. Queenan, J.E. Simpson, V.H. Jennings, E.E Castilla et al. 1998. "Unintended Pregnancies in Women Delivering at 18 South American Hospitals." Human Reproduction 13: 1991-95.

Hellerstedt, W.L., P. Pirie, H. Lando, S. Curry, C. McBride, L. Grothaus and J. Nelson. 1998. "Differences in Preconceptional and Prenatal Behaviors in Women with Intended and Unintended Pregnancies." American Journal of Public Health 88: 663-66.

Henshaw, S. 1998. "Unintended Pregnancy in the United States." Family Planning Perspectives 30: 24-29 and 46.

Kabir, S.M. 1989. "Causes and Consequences of Unwanted Pregnancy from Asian Women's Perspectives." International Journal of Obstetrics 3: S9-S14.

Khuda, B., N.C. Roy and D.M. Rahman, 1999. "Unmet Contraceptive Need in Bangladesh: Evidence from the 1993/1994 and 1996/97 Demographic and Health Surveys." Asia-Pacific Population Journal 14: 37-51.

Klima, C. S. 1998. "Unintended Pregnancy: Consequences and Solutions for a Worldwide Problem." Journal of Nurse-Midwifery 43: 483-91.

Kroelinger, C.D. and K.S. Oths. 2000. "Partner Support and Pregnancy Wantedness." BIRTH 27: 112-19.

Langer, A. 2002. "Unwanted Pregnancy: Impact on Health and Society in Latin America and the Caribbean." Pan American Journal of Public Health 11: 192-204.

Mbizvo, M.T., M. Bonduelle, S. Chadzuka, G. Lindmark and L. Nystrom. 1997. "Unplanned Pregnancies in Harare: What Are the Social and Sexual Determinants?" Social Science and Medicine 45: 937-42.

Moos, M. K. 2003. "Unintended Pregnancies: A Call for Nursing Action." MCN American Journal of Maternal/Child Nursing 28: 24-30.

NIPORT, Mitra and Associates and ORC Macro. 2001. Bangladesh Demographic and Health Survey 1999-2000. National Institute of Population Research and Training (NIPORT), Mitra and Associates, and ORC Macro, Dhaka, and Calverton.

Okonofua, F.E., C. Odimegwa, H. Ajabor, P.H. Daru and A. Johnson. 1999. "Assessing the Prevalence and Determinants of Unwanted Pregnancy and Induced Abortion in Nigeria." Studies in Family Planning 30: 67-77.

PRAMS GRAM. 1996. Unwanted Pregnancy in Oklahoma. Oklahoma Pregnancy Risk Assessment Monitoring System, Oklahoma State Department of Health, 6.

Pulley, L., L.V. Klerman, H. Tang and B.A. Baker. 2002. "The Extent of Pregnancy Mistiming and its Association with Maternal Characteristics and Behaviors and Pregnancy Outcomes." Perspectives on Sexual and Reproductive Health 34: 206-11.

Rai, N.K. and S.M. Dali. 2002. "Making Pregnancy Safer in South-East Asia." Regional Health Forum 6: 19-28.

Rao, A.J. 2001. "A Holistic Approach to Population Control in India." Journal of Bioscience 26: S421-S423.

Saha, K.B. and U. Chatterjee. 1998. "Reproductive Rights in Contraceptive Practices." Health for Millions. 31-32.

Santelli, J., R. Rochat, K. Hatfield-Timajchy, B. Gilbert, K.Curtis and R. Cabral et al. 2003. "The Measurement and Meaning of Unintended Pregnancy." Perspectives on Sexual and Reproductive Health 35: 94-101.

Sein, T. and U.N. Rafei. 2002. "Reproductive Health: No More Cradles in the Graveyards." Regional Health Forum 6: 1-18.

Taylor, J.S. and H.J. Cabral. 2002. "Are Women with an Unintended Pregnancy Less Likely to Breastfeed?" Journal of Family Practice 51: 431-36.

Thapa, S. and S.M. Padhye. 2001. "Induced Abortion in Urban Nepal." International Family Planning Perspectives 27: 144-47 and 151.

Trussell, J., B. Vaughan and J. Stanford. 1999. "Are All Contraceptive Failures Unintended Pregnancies? Evidence from the 1995 National Survey of Family Growth." Family Planning Perspectives 31: 246-47 and 260.